Multi-stream hidden Markov models and wavelets for vocal tract disease detection and pronunciation correction
Evgeny I. Bovbel, Igor E. Kheidorov, Y.V. Pachkovski · 2003
In this paper a common approach based on multi-stream hidden Markov models and wavelets is developed and applied for the tasks of vocal tract pathology detection and deaf people pronunciation correction. Experiments show the high efficiency of such an approach, which can be used in a wide range of biomedical applications.